Repository navigation
CI: vet does not cover tagged acceptance files, and there is no gofmt gate #388
Description
Activity
Concrete instance found while working on #392, worth folding into this issue's scope rather than filing separately.
test/acceptance/request_test.go:356callstriggerEvent, which is defined in anevent-tagged file. So:go vet -tags request ./... → vet: request_test.go:356:2: undefined: triggerEvent go vet -tags "request event" ./... → cleanIt compiles in CI only because slice 1 happens to run
requestandeventtogether (SLICE1_TAGS="request event"inrun_parallel.sh). Therequesttag has a hidden dependency on theeventtag, and resliceing would break it with no warning.This strengthens the case for the vet-per-tag gate proposed here: a per-tag vet would have caught the coupling when it was introduced, whereas the current slice-level run hides it. Worth vetting each tag individually rather than only the slice combinations, since the slices are a CI-parallelism decision and not a statement about what the tags mean.
Not urgent — nothing is broken today — but it is the kind of latent coupling that surfaces as a confusing CI failure much later, after whoever wrote it has moved on.
(Found by Claude while working on #392, on Phil's behalf.)
Summary
Two gaps remain in CI after #384 added
go vet ./...to the unit-test job.1. vet runs untagged only
go vet ./...does not build files behind build tags, and the acceptance suite has 22 of them (basic,metrics,connection,listen,project_use,manual, ...). So none oftest/acceptance/is vetted on a normal run.go vet -tags basic ./...is clean today. The other 21 are unverified. Two real findings have already come out of that directory this week that vet would likely have caught earlier: two no-opappend(sliceLiteral)calls, and an assertion reading the wrong stream.Needs a decision on shape: a tag matrix, a comma-separated
-tagslist (risks duplicate symbols across tags), or vetting the acceptance package separately per tag.2. No gofmt gate
Nothing checks formatting, which is how two files regressed during the 2026-09-01 work and were only noticed by hand.
This cannot simply be switched on: roughly 40 files on
mainare already unformatted, so a gate fails immediately. It needs a repo-widegofmt -wfirst, which is a large mechanical diff best landed on its own rather than inside a feature branch.Suggested order
gofmt -w, alone, no behaviour changes.gofmt -lto the unit-test job, failing on non-empty output.